Evaluation of Food Fortification with Folic Acid for the Primary Prevention of Neural Tube Defects

Objectives and Methods

THE STUDY WAS A MULTI-SITE POPULATION BASED STUDY CARRIED OUT IN TWO PHASES;THE FIRST phase took place prior to mandatory fortification, from November 1997 to March 1998 and the second phase took place after fortification had been implemented for two years, from November 2000 to March 2001.

Because of the high rates of NTDs in Newfoundland, an urban (St. John's) and a rural (Clarenville, Port Blandford, Random Island area) location in this province were chosen as sites for this study. A third site in southeastern Ontario (counties of Frontenac and Lennox & Addington, including the city of Kingston) was also chosen.

Study objectives

The objectives of this study were:

• To determine knowledge and consumption of folic acid supplements, pre- and post-fortification, in women of childbearing age (19-44 years).
• To determine dietary intake of folate pre- and post-fortification in women of childbearing age and in seniors (>= 65 years).
• To determine blood folate and vitamin B₁₂ status in women of childbearing age and in seniors pre-and post-fortification.
• To determine whether the incidence of NTDs in Newfoundland declined following fortification.

Table 1 shows schematically the framework including objectives and sampling of subjects for this study.

Data collection

Through random telephone surveys in the three sites, non-pregnant women of childbearing age (19-44 years) and who spoke English were recruited. In the initial telephone survey, women were asked about their use of vitamin supplements and knowledge of the importance of folic acid for reducing the risk of NTDs or for fetal development. Questions about likelihood of pregnancy and demographic characteristics were asked in order to assess whether these factors influenced knowledge and behaviour.

Women who completed the initial telephone survey were screened for their eligibility for the dietary and blood assessments. Women who were not taking supplements containing folic acid were eligible to participate. Seniors were recruited in the same manner as the sample of women, but were drawn only from St. John's, Newfoundland. Seniors age 65 or over, not diagnosed with vitamin B₁₂ deficiency or anaemia and not taking vitamin B₁₂ or supplements containing folic acid, were eligible for dietary and blood sample assessments.

In order to determine intakes of naturally occurring folate (the form of the vitamin found naturally in foods) pre and post fortification, and dietary intakes of folic acid (the synthetic form of the vitamin) post fortification, a Willett food frequency dietary questionnaire (Willett et al. 1987) was administered to subjects during an in-person interview by trained personnel. There were some modifications to the questionnaire to include common Newfoundland foods and to ensure that all foods high in folate were included. The dietary questionnaire was used to estimate an average frequency of consumption of 124 food items over the previous period of one year. The Willett food frequency dietary questionnaire is well validated (Willett et al. 1988) and proved easy to administer for this sample population.

The women and senior participants were also asked to provide a sample of blood in order to determine blood folate and vitamin B₁₂ status, pre- and post- fortification. Laboratory tests for complete blood count (CBC), red blood cell (RBC) folate, serum folate, creatinine, vitamin B₁₂, plasma homocysteine (HCY) and methylmalonic acid (MMA) were conducted at the laboratories of the Health Care Corporation of St. John's.

In order to examine temporal changes in the incidence of NTDs in Newfoundland, data for 1976 to 2001 were compiled by the Newfoundland and Labrador Medical Genetics Program. NTD cases include anencephaly, spina bifida and encephalocele diagnosed in live births, stillbirths and fetuses from pregnancies terminated after a prenatal diagnosis of an NTD.

Data analysis

Data collected from these sites were compared between Phase I and Phase II. Data from the blood analyses were tested for normality with the Komogorov-Smirnov test, and the Shapiro-Walk test when the number of data points was less than 50. Differences between groups were tested using the non-parametric Kruskal-Wallis test and the Mann-Whitney U test or Student's t-test. The distributions of plasma MMA, plasma HCY, serum folate, RBC folate and serum vitamin B₁₂ were skewed. Values were therefore log transformed to give an approximate normal distribution for estimation of geometric mean and confidence intervals. Unless otherwise stated, all laboratory values presented in this paper are geometric means and 95% confidence intervals (CI). Differences in the frequency of high or low results based on reference values were tested by Pearson chi-square statistics.

Incidence of NTDs was defined as the number of above described NTD cases, divided by the total number of live births, stillbirths, and pregnancy terminations for an NTD (termed as "births" hereafter). First we examined the temporal trend in annual incidence of NTDs from 1976 to 2001 using 3-year moving average rates, then we focused on comparison of the incidence data for the most recent 11 years, identified as pre-supplementation (1991-1993), pre-fortification (1994-1997) and post-fortification (1998-2001). We regard the year 1997 as a transition period, or partial fortification period, since fortification of white flour and enriched pasta and cornmeal was permitted in Canada as of December 1996 (Canada Gazette 1996). Thus we also analysed the incidence data using 1994-1996 as a pre-fortification period.

All data for this study were entered into SPSS (the Statistical Package for Social Sciences) Rel. 10.0 after the end of each phase. Data from the dietary interviews were analyzed using Epi-Info (Version 6.04d), while the laboratory data and the data about knowledge and use of supplements were analyzed using SPSS.